A molten slag state recognition method based on furnace window flame image and multi-physical field simulation

By constructing a multiphysics coupled simulation model and parameterized slag permeability, combined with flame image data, the problem of real-time and accurate identification of slag state in converter steelmaking was solved, realizing the analysis of non-uniformity of slag state and accurate location of abnormal areas, supporting intelligent steelmaking.

CN122154541APending Publication Date: 2026-06-05SICHUAN UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SICHUAN UNIV
Filing Date
2026-02-28
Publication Date
2026-06-05

AI Technical Summary

Technical Problem

Existing technologies cannot achieve real-time, accurate, and mechanistic perception of slag state during converter steelmaking. They lack multi-physics coupling models, cannot accurately explain the impact of slag state changes on flame dynamic characteristics, and lack data-mechanism closed-loop verification, resulting in inaccurate state diagnosis.

Method used

A multiphysics coupled simulation model is constructed. By parameterizing the permeability of molten slag and using it as a boundary condition, and combining it with flame image data from industrial sites, a quantitative correlation between molten slag state and flame characteristics is established, enabling the identification and accurate prediction of the non-uniformity of molten slag state.

Benefits of technology

It achieves highly reliable and accurate online identification of slag condition, can locate local abnormal areas, supports intelligent steelmaking decision-making, reduces implementation costs, and has expansion potential.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122154541A_ABST
    Figure CN122154541A_ABST
Patent Text Reader

Abstract

The application discloses a molten slag state recognition method based on a furnace window flame image and multi-physical field simulation, and comprises the following steps: 1, a converter geometric model and a multi-physical field coupling mathematical model are constructed; 2, a parameterized representation of molten slag permeability is performed, and a dynamic boundary condition is set; 3, simulation is performed, and flame characteristic quantitative indexes are extracted; 4, industrial field flame images are collected and processed; 5, image characteristic inversion and contrast verification are performed; and 6, a permeability state recognition model is established. The application establishes a quantitative physical correlation between molten slag permeability and furnace window flame characteristics, realizes quantitative recognition of molten slag state space non-uniformity, and constructs a reliable verification closed loop of a physical mechanism-numerical simulation-industrial data.
Need to check novelty before this filing date? Find Prior Art